JPH04353765A - Detecting device for magnetic variation - Google Patents

Abstract

PURPOSE:To obtain a detecting device of magnetic variation which can execute highly-precise detection even when there are various factors of changes such as a change of a gap between a body to be detected and a sensor, nonuniformity in an output of the sensor itself and the temperature characteristic thereof. CONSTITUTION:The present value Dn of a digital output is taken in (S1), the update maximum value Dmax and minimum value Dmin are determined and a hysteresis width H is computed therefrom (S2). When a digital output Di shows a tendency to increase (S3), a binary-coded output Vout is switched over to '1' only when Dn > (Dmin+H) is satisfied (S4, S5). When the judgement at S3 shows a tendency to decrease, on the other hand, the binary-coded output Vout is switched over to '0' only when Dn < (Dmax-H) is satisfied (S6, S7). Since the hysteresis width is updated by using the update maximum/ minimum value, the same angular phase can always be made the time of inversion of the binary-coded output Vout.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] The present invention relates to a magnetic change repeating state detecting means for detecting a repeating state of magnetic change accompanying the movement of a detected object, and an increase in the magnetic change detected by the magnetic change repeating state detecting means. Alternatively, the present invention relates to a magnetic change detection device including switching timing determining means for determining that the magnetic change has switched between an increasing tendency and a decreasing tendency when the decreasing tendency exceeds a predetermined reference value.

0002

2. Description of the Related Art Such magnetic change detection devices have been used in various ways, for example, in devices that detect the state of linear motion or rotational motion of a detected object by utilizing changes in the output voltage of a Hall element. The output voltage change of the Hall element is
AD converted based on a predetermined binary judgment algorithm,
It is given as a binary output and used for various calculations.

As shown in FIG. 8, the conventional binary judgment algorithm judges the magnitude relationship between the current value Dn of the output voltage from the Hall element and the reference value Do (S101), and then determines the current value Dn.
If it is larger than the standard value Do, it is judged that there is an increasing trend and 2
The set value Qn for value output is set to "1" (S102), and the other values are judged to be decreasing, and the set value Qn for binary output is set to "1" (S102).
0" (S103), and only when the absolute value of the difference between the current value Dn and the reference value Do exceeds the predetermined hysteresis width H, the binary output is performed based on the previously set binary output setting value Qn. Vout is performed and the reference value Do is reset to the current value Dn at that time (S104, S105).

0004

[Problems to be Solved by the Invention] However, in the case of rotation detection, for example, variations in the gap between the sensor and the rotating body,
The maximum and minimum values fluctuate due to variations in Hall element output. In contrast, conventional binary determination algorithms do not take such fluctuations into consideration.

Therefore, as is clear from FIG. 9, at time t1
At times t103 and t104, the output switches approximately halfway between the maximum and minimum values, but as the maximum/minimum width increases, the hysteresis width H changes when approaching the maximum or minimum value, such as at times t103 and t104. is reached, and the output is switched. For this reason, the period (t103
-t104) is shorter than the period (t101 -t102). On the other hand, when the maximum and minimum width decreases, as is clear from the period (t105 - t106),
The switching timing of the binary output Vout becomes long. As described above, in the conventional device, there was a problem in that the switching timing of the binary output fluctuated and the detection accuracy deteriorated.

Therefore, we have developed a magnetic change detection device that can perform highly accurate detection even when there are various fluctuation factors such as variations in the gap between the detected object and the sensor, variations in the output of the sensor itself, and temperature characteristics. The present invention has been completed for the purpose of providing.

[0007]

[Means and operations for solving the problems] In order to achieve the above object, the magnetic change detection device of the present invention provides a magnetic change repeated state detection means for detecting a repeated state of magnetic change accompanying the movement of a detected object. and a switching timing at which it is determined that the magnetic change has switched between an increasing tendency and a decreasing tendency when the increasing or decreasing tendency of the magnetic change detected by the magnetic change repeated state detection means exceeds a predetermined reference value. a magnetic change detection device comprising: a immediately preceding peak value extracting means for extracting a maximum value and a minimum value of a value detected by the magnetic change repetition state detection means in a magnetic change repetition state immediately before a switching timing determination; The present invention is characterized by comprising a reference value updating means for updating the reference value based on the maximum value and minimum value extracted by the immediately preceding peak value extraction means.

[0008] According to the magnetic change detection device of the present invention, the reference value as a judgment criterion in the switching timing judgment means is determined based on the immediately previous magnetic change repeating state by providing the immediately preceding peak value extraction means and the reference value updating means. The switching timing and the maximum and minimum values are updated based on the maximum and minimum values of the detected values at The relationship with the minimum value can be maintained constant, and high detection accuracy can be obtained.

[0009]

[Embodiment] Next, an embodiment to which the present invention is applied will be described. The embodiment is applied to a rotational speed detection device, and as shown in FIG. 3, a magnet 20 for magnetic field bias is arranged so as to face teeth 10a made of magnetic material of a rotor 10, and a magnet 20 is placed between them. , a molded IC 40 in which a magnetic detection circuit 50 including a Hall element is sealed is arranged. In addition, 3 in the figure
0 is a case made of non-magnetic material, 41 is an input/output pin, 42
is the substrate.

As shown in FIG. 4, this magnetic detection circuit 50 includes a first voltage/frequency conversion circuit VCO1 made of, for example, a CMOS type voltage controlled oscillation circuit, and a second voltage/frequency conversion circuit VCO1 also made of a CMOS type voltage controlled oscillation circuit. /Frequency conversion circuit VC
O2 in parallel, each voltage/frequency conversion circuit VCO1
, the Hall voltage V from the Hall element HE with respect to VCO2
P and VM are applied. Therefore, each of the voltage/frequency conversion circuits VCO1 and VCO2 generates oscillation outputs of frequencies f1 and f2, respectively, in response to changes in the Hall voltages VP and VM from the Hall element HE. These two oscillation outputs are input to a frequency difference digitization circuit 51. This frequency difference digitization circuit 51 digitizes the difference f1-f2 between frequencies f1 and f2 and sends the digitized output Ni to a digital filter 52 at the subsequent stage.
give to The digital filter 52 outputs this numerical value N
i is converted into a digital output Di, which is further provided to a subsequent comparator 53. This comparator 53 executes a binarization process based on the digital output Di, and the binarization output V
Output out. This binary output Vout allows the change in magnetic field strength due to the rotation of the rotor 10 to be quantitatively understood. FIG. 5 schematically shows the relationship between magnetic field strength and each output.

[0011] Note that the voltage/frequency conversion circuit VCO1, V
CO2, frequency difference digitization circuit 51, digital filter 5
2 and the comparator 53, relative operation timings and the like are adjusted by a clock signal CLK. The binarization process performed by the comparator 53 is based on an algorithm as shown in FIG.

First, the current value Dn of the digital output from the digital filter 52 is taken in (S1). Next, the latest maximum value Dmax and minimum value Dmin are determined, and the hysteresis width H is calculated based on them (S2). Thereafter, it is determined whether the digital output Di is increasing or decreasing (S3).

[0013] If this judgment result shows an increasing tendency, the current value Dn is smaller than the reference value Do (Dmin) by the hysteresis width H.
It is determined whether or not the amount has exceeded the limit (S4). And Dn>
(Dmin+H), then the binary output Vout
is switched to "1" (S5). On the other hand, if the judgment in S3 indicates a decreasing trend, the current value Dn becomes the reference value Do(Dma
x) is below the hysteresis width H (
S6). Then, only when Dn<(Dmax -H) is satisfied, the binarized output Vout is switched to "0" (
S7).

[0014] In the process of S2, the latest maximum value Dmax
, the minimum value Dmin can be obtained, for example, as follows. First, after setting the initial value to Dn=Dmax=Dmin, every time Dn>Dmax, "Dm
ax←Dn”, and conversely, each time Dn<Dmin, it is updated as “Dmin←Dn”. In this way, it is possible to know the maximum and minimum actual detected values. However, this value cannot be considered as the latest value. Therefore, in the process of S5, the binarized output Vout is set to "1".
At the same time, the maximum value Dmax is cleared, and S7
In the process, the binarized output Vout may be switched to "0" and the minimum value Dmin may be cleared.

As a result, as shown in FIG. 2, from time t1 to
For the maximum value Dmax used to calculate the hysteresis width H during t2, max1, which is the value of the immediately preceding peak, is used, and min1, which is the value of the immediately preceding trough, is used for the minimum value Dmin.
will be used. Note that the calculation formula for the hysteresis width H uses the calculation coefficient m, for example, H=m(D
max + Dmin). Therefore, H1
=m(max1+min1) as the hysteresis width,
A determination is made based on this, and the binarized output Vout is switched from "1" to "0" at time t2.

At this time, the minimum value Dmin is cleared, so after time t2, the maximum value Dmax is max
Although it remains at 1, the minimum value Dmin becomes the current value Dn and gradually decreases until time t3. Note that from time t2 to time t3 when the minimum value Dmin is determined.
Until then, H1 is held.

Since the second valley is crossed at time t3, the minimum value Dmin is thereafter determined to be the value min2 of the second valley. The hysteresis width is updated only when new maximum and minimum values are determined. Therefore, the hysteresis width is H2=m(max1+min2)
Therefore, in the period after time t3, the binarized output Vout is set as "
The timing for switching from "0" to "1" is determined. As a result, switching of the binarized output Vout is executed at time t4.

Then, the maximum value Dmax is cleared and gradually increases as the current value Dn increases,
It is determined at time t5 when the next mountain is crossed. As a result,
Maximum value Dmax = max2, minimum value Dmin = mi
Hysteresis width H3=m(max2+min2
) is calculated, and the same process is repeated.

The comparator 53 has a detailed circuit configuration as shown in FIG. 6, and is configured to realize this binarization algorithm. The digital output Di input from the digital filter 52 is the current value latch circuit 5.
3a, a reference value determination and latch circuit 53b, a maximum value detection circuit 53c, and a minimum value detection circuit 53d.

The current value latch circuit 53a outputs the current value Dn to the subsequent difference detection circuit 53e. Difference detection circuit 5
3e is the current value Dn and the reference value judgment and latch circuit 5.
The difference Sn from the reference value Do output from 3b is determined and output to the subsequent hysteresis calculation circuit 53f. In addition, the difference detection circuit 53e also outputs an increase/decrease determination signal Sn (MSB) by a latch circuit 53g, a reference value determination and latch circuit 53b, and a hysteresis width setting circuit 53h.
Output to. Incidentally, the increase/decrease determination signal Sn (MSB) is D
This is a signal that instructs high level output when n-Do is positive, and low level output when n-Do is negative.

On the other hand, the maximum value detection circuit 53c calculates the maximum value Dmax by updating the input digital output Di when it exceeds the maximum value detected so far.
It is output to the subsequent hysteresis width setting circuit 53h. Similarly, the minimum value detection circuit 53d also updates the minimum value Dmin and outputs it to the hysteresis width setting circuit 53h.

The hysteresis width setting circuit 53h calculates a hysteresis width H=m(Dmax+Dmin) based on the input maximum value Dmax and minimum value Dmin, and outputs this to the hysteresis calculation circuit 53f. The hysteresis calculation circuit 53f is a difference detection circuit 53e.
The difference Sn given from the hysteresis width H is compared with the hysteresis width H, and if the difference Sn exceeds the hysteresis width H, a latch signal is output to the reference value determination and latch circuit 53b and the output latch circuit 53g.

When the reference value determination and latch circuit 53b receives the latch signal from the hysteresis calculation circuit 53f, it clears the reference value Do and outputs the digital output Di (ie, current value Dn) input from the digital filter 52. The reference value Do is set and fixed at the time when an extreme value appears. Therefore, to explain with reference to FIG. 2 shown above, the reference value Do is reset at time t2 and becomes reference value Do=Dn(t2) until time t3, and between time t3 and t4, reference value Do=Dn(t2). min2, and time t4 to t5
During this period, the reference value Do=Dn(t4), and after time t5, the reference value Do=max2.

On the other hand, when the latch signal is output from the hysteresis calculation circuit 53f, the output latch circuit 53g binarizes and outputs a high level signal or a low level signal based on the increase/decrease determination signal Sn (MSB) from the difference detection circuit 53e. Output as Vout. Note that the current value latch circuit 5
3a, reference value determination and latch circuit 53b, maximum value detection circuit 53c, minimum value detection circuit 53d, difference detection circuit 53e
, the hysteresis calculation circuit 53f, the hysteresis width setting circuit 53h, and the like are synchronized by a clock signal CLK.

Next, a description will be given of FIG. 7 which schematically shows the difference in operation and effect between the binarization determination algorithm shown in FIG. 8 described as a conventional example and the binarization determination algorithm of this embodiment. FIG. 7A shows that the gap between the teeth 10a of the rotor 10 and the magnets 20 is wide and the output voltage V from the Hall element HE is
When P and VM are small, the digital filter 5
The digital output Di output from 2 is on the upper stage, and conversely, the gap is narrow and the output voltages VP and V from the Hall elements HE are
The digital output Di output from the digital filter 52 in a state where M is large is shown in the lower row. In addition,
In the figure, ts is the sampling period, 1/ts=f1
- Can be expressed as n (where n is the frequency division ratio and f1 is the frequency of the VCO1).

Now, assume that the binarized output Vout is switched at the time when the current value Dn becomes halfway between the peak and valley of the digital output Di. When the binarization determination algorithm of this embodiment is used, the hysteresis width is updated using the latest maximum value and minimum value. Therefore, when considering the reversal timing when the binarized output is inverted between the maximum value and the minimum value, if the Hall output is small, the reversal time is the time when the digital value Di becomes "5". becomes. On the other hand, when the Hall output is large, the digital value Di is “1”.
The time when the value becomes 0 is the reversal time. Therefore, as is clear from the figure, the error in the reversal timing of the binarized output due to the magnitude of the Hall output is only the angular phase difference tb, as shown in (B) in the figure. On the other hand, in the conventional algorithm, the hysteresis width is fixed, so whether the Hall output is small or the Hall output is large, the digital value is, for example, "3".
Since the time when the value becomes , is the inversion time of the binarized output, the error becomes larger than the angular phase difference ta.

In this way, when the Hall output fluctuates for some reason, the switching timing of the binarized output changes greatly in the conventional example, but according to this embodiment, the change in the Hall output The binarized output can always be switched at a constant angular phase without being influenced by the angular phase.

As a result, the precision and reliability of various calculations using this binarized output can be improved. Although one embodiment of the present invention has been described above, the present invention is not limited to this.
For example, it can be configured as a magnetic change detection device for detecting displacement or linear motion of an object to be detected, and various embodiments can be adopted within the scope of the present invention even if it is a specific circuit configuration. can.

[0029]

As described above, according to the magnetic change detection device of the present invention, by adopting a unique algorithm, fluctuations in the gap between the detected object and the sensor, variations in the output of the sensor itself, and temperature can be detected. Even if there are various fluctuation factors such as characteristics, highly accurate detection can be performed.

[Brief explanation of drawings]

FIG. 1 is a flowchart of a binarization determination algorithm according to an embodiment.

FIG. 2 is a timing chart explaining the effect.

FIG. 3 is a schematic configuration diagram of a rotational speed detection device according to an embodiment.

FIG. 4 is a configuration diagram of a magnetic change detection circuit employed therein.

FIG. 5 is a timing chart showing the relationship between magnetic field strength and each output in this magnetic change detection circuit.

FIG. 6 is a detailed circuit configuration diagram of a comparator of the magnetic change detection circuit of the embodiment.

FIG. 7 is an explanatory diagram comparing the operation and effectiveness of the embodiment and the conventional example.

FIG. 8 is a flowchart of a conventional binarization determination algorithm.

FIG. 9 is a timing chart showing the problem.

[Explanation of symbols]

10...Rotor, 10a...Teeth, 20...Magnet,
30...Case, 40...Mold IC, 50...
- Magnetic detection circuit, 51... Frequency difference digitization circuit, 52
...Digital filter, 53...Comparator, 5
3a...Current value latch circuit, 53b...Reference value judgment and latch circuit, 53c...Maximum value detection circuit, 53d
...Minimum value detection circuit, 53e...Difference detection circuit, 5
3f...Hysteresis calculation circuit, 53g...Output latch circuit, 53h...Hysteresis width setting circuit, HE
...Hall element, VCO1, VCO2...Voltage/frequency conversion circuit.

Claims (1)

[Claims] 1. A repeated magnetic change state detection means for detecting a repeated state of magnetic change accompanying the movement of a detected object, and a magnetic change state detection means that detects a repeated state of magnetic change caused by the movement of a detected object, and a magnetic change state detection means that detects a repeated magnetic change state detecting means, wherein a tendency of increase or decrease in the magnetic change detected by the repeated magnetic change state detection means is determined by a predetermined tendency. In a magnetic change detecting device, the magnetic change detecting device is provided with a switching timing determining means for determining that the magnetic change has switched between an increasing tendency and a decreasing tendency when the magnetic change exceeds a reference value. Immediate peak value extraction means for extracting the maximum value and minimum value of the detected value by the magnetic change repeated state detection means;
A magnetic change detection device comprising: reference value updating means for updating the reference value based on the maximum value and minimum value extracted by the immediately preceding peak value extraction means.